line trees, incidence of infection was actually approximately 12 percent less 

 than for the line trees. Even for trees attacked by the beetle in 1937, in 

 which a whole year had elapsed during which fungal infection might be pre- 

 sumed to have increased and extended rapidly, incidence of infection was 

 only 19 percent greater than for the line trees. These results give no indica- 

 tion that root infection increases the likelihood of beetle attack. 



The results further indicate that infection may take place and spread 

 rapidly following beetle attack. This possibility is supported by observa- 

 tion of extensive, apparently fresh, mycelial fans under the bark in above- 

 ground parts of dying beetle-attacked trees. Of the 1938-attacked trees, 

 13 percent on Deception Creek and 18 percent on Kelley Creek had half or more 

 than half of the circumference of root collar killed by root infection. 

 For the 1937-attacked trees, the figures are 31 and 50 percent, respectively. 

 The 1938-attacked trees on all the areas appear to be comparable with the 

 line trees, of which 16 percent had half or more of the circumference killed. 

 As might be expected, degree of root infection parallels incidence of infec- 

 tion, being greater a year after beetle attack than in the same season. 



2. Relation to Possibly Influencing Factors 



Since the number of infected trees examined was small, especially those 

 attacked by the beetle in 1938, trees examined in all drainages were combined 

 by years of beetle attack, and incidence of infection was determined for dif- 

 ferent classes of site quality, tree height, site moisture, and aspect. 

 The results of tabulating incidence of infection in each of three site-quality 

 classes substantiate the results with line trees and indicate that infection 

 is more frequent on poorer sites than on good sites. As might be expected, 

 incidence is higher in trees attacked in 1937 than in 1938 for all site classes, 

 but the relationship is similar for both groups of trees. When incidence of 

 infection was tabulated on the basis of tree height, comparable results were 

 obtained and the results with the line trees were substantiated. The results 

 of tabulating incidence of infection in each of three site-moisture classes 

 indicate that the trees attacked by the beetle in 1938 substantiate the 

 results with line trees and show that on wet sites trees are less subject to 

 infection than on drier sites. The results with trees attacked by the beetle 

 in 1937 appear to be opposite but are probably indeterminate. Since the 

 beetle-attacked trees covered a wider range of aspects than did the line 

 trees, an analysis was made to see if they might give an indication of a 

 possible relationship between incidence of infection and aspect. The results 

 were inconclusive, however, save to suggest that, as might be expected, in- 

 cidence of infection is somewhat greater on the drier aspects (west, south, 

 and ridge top) than on the wetter aspects (bottom, north, and east). 



-9- 



